Locomotion is the product of neural output acting on muscles driving a mechanically complex body in an unpredictable environment. The lamprey is a simple, well-studied and relatively tractable vertebrate model with which to probe this neuromechanical system. We hypothesize that steady locomotion in a predictable environment requires only the central pattern generator (CPG) without the necessity of other input. We also hypothesize that in an unpredictable environment sensory feedback combined with strong intersegmental coupling is necessary. To investigate these hypotheses we will develop an integrated model, of lamprey swimming LAMPREYCOMP with Thelma Williams, from London, and researchers Philip Holmes and Alexander Smits from Princeton University. The model spans CPG, sensory feedback, muscle mechanics, body mechanics, and fluid mechanics, and is a full model of a complex behavior in a vertebrate, albeit a simple one. Cohen and colleagues will perform experimental studies of the CPG and its response to sensory feedback from spinal mechanoreceptors, skin and lateral line and will develop the component of LAMPREYCOMP that maps sensory input to motor nerve output. Work on muscle and body mechanics will be done by Holmes and Williams. Smits and Holmes will study the fluid mechanics of the swimming animals both experimentally and theoretically. The experimental work will include developing a mechanical analog, P-RAY, whose motion will be adjusted to reproduce that of live animals, allowing measurements of fluid motion and pressure variations along the body. The theoretical work will include developing the fluid dynamical component of LAMPREYCOMP, in consultation with C. Peskin, L. Fauci and colleagues. LAMPREYCOMP will be tested to insure that it reproduces the behavior of whole animal swimming, various reduced preparations and P-RAY. It will then be used to investigate the effect of manipulations, such as removing sensory feedback from spinal mechanoreceptors that are not experimentally possible. We also show that all components of the protocol are possible. Because the lamprey is a model system for all of vertebrate locomotion, our hypotheses and models will have broad implications for more advanced organisms in which such a complete approach is not presently feasible.